| 摘要: |
| 知识图谱是一种用网络结构存储知识的知识库, 在知识图谱中, 单条知识被表示成三元组的形式, 即(头实体, 关系, 尾实体). 得力于知识图谱在各个领域的广泛应用, 面向知识图谱的图嵌入学习也得到越来越多研究人员的关注. 面向知识图谱的图嵌入学习任务旨在为图谱中的实体与关系学习低维且稠密的向量, 通过图嵌入向量表达实体与关系的语义信息以及度量实体之间、关系之间、实体与关系之间的语义联系, 已有许多研究证明图嵌入模型在下游任务中的有效性. 近年来, 越来越多研究人员开始关注知识图谱的图嵌入学习, 并取得大量的研究成果, 尝试将图嵌入算法分成了基于转移思想、基于张量分解、基于传统深度学习模型、基于图神经网络以及融入额外信息的图嵌入学习共5大类, 梳理、介绍各类图嵌入算法的设计思路、算法特征以及优缺点, 以帮助指导初步接触该领域的研究人员快速学习了解该研究领域的相关模型和算法. |
| 关键词: 知识图谱 图嵌入学习 表示学习 链接预测 |
| DOI：10.13328/j.cnki.jos.006426 |
| 分类号: |
| 基金项目:国家自然科学基金(61772157, 61832003, U1866602) |
|
| Survey on Knowledge Graph Embedding Learning |
|
YANG Dong-Hua1,2, HE Tao1, WANG Hong-Zhi1, WANG Jin-Bao1
|
|
1.Faculty of Computing, Harbin Institute of Technology, Harbin 150001, China;2.Center of Analysis, Measurement and Computing, Harbin Institute of Technology, Harbin 150001, China
|
| Abstract: |
| Knowledge graphs (KGs) serve as a kind of knowledge base by storing facts with network structure, representing each piece of fact as a triple, i.e. (head, relation, tail). Thanks to the general applications of KGs in various of fields, the embedding learning of knowledge graph has also quickly gained massive attention. This study tries to classify the existing embedding algorithms as five types: translation-based models, tensor factorization-based models, traditional deep learning-based models, graph neural network-based models, and models by fusing extra information. Then, the key ideas, algorithm features, advantages and disadvantages of different embedding models are introduced and analyzed to give the first-time researchers a guideline that can be referenced to help researchers quickly get started. |
| Key words: knowledge graph (KG) graph embedding learning representation learning link prediction |
